Map construction method and device, equipment and storage medium

Abstract

The invention discloses a map construction method, device and equipment and a storage medium, and relates to the technical field of automatic driving, and the method comprises the following steps: obtaining navigation route information of a target vehicle; according to the navigation route information, a driving-out lane and a driving-in lane of the target vehicle are obtained through matching in a pre-constructed lane map, the lane map is constructed according to driving environment information within a preset range of the position where the target vehicle is located, and the driving environment information comprises boundary lines of all lanes; according to the method, road connection processing is carried out on the driving-out lane and the driving-in lane to obtain a target lane level map, a lane map can be generated only through driving environment information, the driving-out lane and the driving-in lane are determined in the lane map according to navigation route information, then road connection processing is carried out, and therefore the accurate lane level map is constructed. The accuracy and stability of lane-level map construction are improved on the premise that no high-precision map dependence exists, and therefore the safety and scene universality of an automatic driving system are improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a map building method, apparatus, device, and storage medium. Background Technology

[0002] Autonomous driving technology has high requirements for road environment perception and map building, as well as high precision, real-time performance and dynamic adaptability. The current technical path based on high-precision maps has problems such as high data collection and update costs, long cycles, poor adaptability to dynamic scenarios and limited coverage. The existing mapless or map-sparse mode cannot build accurate and stable real-time lane-level maps, which is not conducive to improving the safety and scenario universality of autonomous driving systems.

[0003] Therefore, it is necessary to propose a solution to improve the accuracy and stability of lane-level map construction without relying on high-precision maps.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a map building method, apparatus, device, and storage medium, which aims to improve the accuracy and stability of lane-level map building without relying on high-precision maps, thereby enhancing the safety and scenario applicability of autonomous driving systems.

[0006] To achieve the above objectives, this application provides a map construction method, the method comprising: Obtain navigation route information of the target vehicle, wherein the navigation route information is the preset driving path information of the target vehicle; Based on the navigation route information, the exit lane and entry lane of the target vehicle are matched in a pre-constructed lane map. The lane map is constructed based on driving environment information within a preset range of the target vehicle's location. The driving environment information includes the boundary lines of each lane. The exit lane includes the lanes that the target vehicle can currently exit from. The entry lane includes the lanes that the target vehicle can enter. The exit lane and the entry lane are connected to form a road connection to obtain a target lane-level map.

[0007] In one embodiment, the lane map is constructed according to the following steps: Obtain driving environment information within a preset range of the target vehicle's location; Based on the lane boundary lines in the driving environment information, rasterized data is generated; The rasterized data is extracted to generate a Venn diagram; The Vinograph is refined and its branch points are merged to obtain the lane map.

[0008] In one embodiment, the step of refining the Vinio map and merging branch points to obtain the lane map includes: The Venn diagram is refined to generate lane centerlines; and the branch points in the Venn diagram are merged to generate at least one intersection point. The lane map is constructed based on the lane centerline and the at least one intersection point.

[0009] In one embodiment, the at least one intersection point includes a first type of intersection point and / or a second type of intersection point, and the step of merging the branch points in the Venn diagram to generate at least one intersection point includes: The degree of each branch point is determined based on the number of connected edges at each branch point in the Vinograph. Designate the branch points with a degree of 1 as leaf nodes and the branch points with a degree greater than 1 as non-leaf nodes. Adjacent points whose distance to the leaf node is less than a first preset distance threshold are merged to obtain the first type of intersection point; and adjacent points whose distance to the non-leaf node is less than a second preset distance threshold are merged to obtain the second type of intersection point.

[0010] In one embodiment, the step of matching the exit lane and entry lane of the target vehicle in a pre-built lane map based on the navigation route information includes: The navigation route information is matched with the lane map by at least one of geometric matching, lane attribute matching, traffic condition matching, and dynamic environment matching to obtain the exit lane and the entry lane. The geometric matching includes matching the user's location in the lane map with the lane location in the navigation route information; the lane attribute matching includes matching the lane type in the lane map with the lane type in the navigation route information; the traffic condition matching filters the lanes in the lane map according to the traffic conditions in the navigation route information; and the dynamic environment matching filters the lanes in the lane map according to the real-time traffic information in the navigation route information.

[0011] In one embodiment, the step of performing road connection processing on the exit lane and the entry lane to obtain a target lane-level map includes: Based on the lane map, determine the lane center lines corresponding to the exit lane and the entry lane respectively; The center lines of the lanes corresponding to the exit lane and the entry lane are smoothed and extended, and lane attributes are added to generate the processed entry lane and exit lane. The lane attributes of the entry lane include the boundary line forming the entry lane and / or the lanes adjacent to the entry lane. The lane attributes of the exit lane include the boundary line forming the exit lane and / or the lanes adjacent to the exit lane. The processed inbound and outbound lanes are then connected to form a road connection to obtain the target lane-level map.

[0012] In one embodiment, the step of performing road connection processing on the processed inbound and outbound lanes to obtain the target lane-level map includes: Extract the lane lines corresponding to the processed inbound and outbound lanes; The lane lines to be connected are selected from the lane lines corresponding to the processed exit lanes. The lane lines to be connected and the lane lines corresponding to the processed entering lanes are topologically connected to obtain the target lane-level map.

[0013] In one embodiment, the step of selecting the lane line to be connected from the lane lines corresponding to the processed exit lane includes: When there is one driving lane in the processed exit lane, the lane line corresponding to the processed exit lane is used as the lane line to be connected. When there are at least two driving lanes in the processed exit lane, determine the lane type of each driving lane; If there is a driving lane of the type of motor vehicle lane, then determine the driving direction indicator associated with the driving lane of the type of motor vehicle lane, wherein the driving direction indicator is used to indicate the direction in which the target vehicle can travel in the driving lane; Based on the driving direction indicator, the lane line to be connected is selected from the lane lines corresponding to the driving lanes belonging to the motor vehicle lane type. Furthermore, to achieve the above objectives, this application also proposes a map building apparatus, which includes: The acquisition module is used to acquire navigation route information of the target vehicle, wherein the navigation route information is the preset driving path information of the target vehicle; The matching module is used to match the exit lane and entry lane of the target vehicle in a pre-built lane map based on the navigation route information. The lane map is constructed based on driving environment information within a preset range of the target vehicle's location. The driving environment information includes the boundary lines of each lane. The exit lane includes the lanes that the target vehicle can currently exit from. The entry lane includes the lanes that the target vehicle can enter. The connection module is used to perform road connection processing on the exit lane and the entry lane to obtain a target lane-level map.

[0014] In addition, to achieve the above objectives, this application also proposes a map building device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the map building method as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and which, when executed by a processor, implements the steps of the map construction method described above.

[0016] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the map construction method described above.

[0017] One or more technical solutions proposed in this application have at least the following technical effects: By acquiring the navigation route information of the target vehicle, wherein the navigation route information is the target vehicle's preset driving path information; based on the navigation route information, the exit lane and entry lane of the target vehicle are matched in a pre-constructed lane map, wherein the lane map is constructed based on driving environment information within a preset range of the target vehicle's location, the driving environment information including the boundary lines of each lane; the exit lane includes the lanes that the target vehicle can currently exit from; the entry lane includes the lanes that the target vehicle can enter; road connection processing is performed on the exit lane and the entry lane to obtain a target lane-level map. The lane map can be generated with only driving environment information. The exit lane and entry lane are determined in the lane map based on the navigation route information, and then road connection processing is performed, thereby constructing an accurate lane-level map. This improves the accuracy and stability of lane-level map construction without relying on high-precision maps, thereby enhancing the safety and scenario versatility of the autonomous driving system. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an embodiment of the map construction method of this application. Figure 2 This is a schematic diagram illustrating the overall process according to an embodiment of this application; Figure 3 This is a flowchart illustrating Embodiment 2 of the map construction method of this application; Figure 4 This is a schematic diagram of a Vinio diagram according to an embodiment of this application; Figure 5 This is a schematic diagram of the intersection point according to an embodiment of this application; Figure 6 This is a flowchart illustrating Embodiment 3 of the map construction method of this application; Figure 7 This is a schematic diagram illustrating driving environment information according to an embodiment of this application; Figure 8 This is a schematic diagram of a lane map according to an embodiment of this application; Figure 9 This is a schematic diagram illustrating a left-turn scenario according to an embodiment of this application; Figure 10 This is a schematic diagram of a target lane-level map according to an embodiment of this application; Figure 11 This is a schematic diagram of the module structure of the map building device according to an embodiment of this application; Figure 12 This is a schematic diagram of the device structure of the hardware operating environment involved in the map construction method in this application embodiment.

[0021] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of this application embodiment is as follows: By acquiring the navigation route information of the target vehicle, wherein the navigation route information is the preset driving path information of the target vehicle; based on the navigation route information, matching the exit lane and entry lane of the target vehicle in a pre-constructed lane map, wherein the lane map is constructed based on driving environment information within a preset range of the target vehicle's location, the driving environment information including the boundary lines of each lane; the exit lane includes the lanes that the target vehicle can currently exit; the entry lane includes the lanes that the target vehicle can enter; road connection processing is performed on the exit lane and the entry lane to obtain a target lane-level map. Only driving environment information is needed to generate the lane map. The exit lane and entry lane are determined in the lane map based on the navigation route information, and then road connection processing is performed, thereby constructing an accurate lane-level map. This improves the accuracy and stability of lane-level map construction without relying on high-precision maps, thereby enhancing the safety and scenario applicability of the autonomous driving system.

[0025] In this embodiment, for ease of description, the following description will focus on the map building device as the executing entity.

[0026] Technical terms used in the embodiments of this application: A Voronoi diagram, also known as a Thiessen polygon, is a graphical representation of space divided into discrete regions. It takes a set of points as input and divides the space into multiple regions based on the distances between these points. Each region surrounds one input point, with the nearest point being the furthest from points in other regions.

[0027] Voronoi diagrams offer flexibility and scalability in meshing, enabling the generation of various types of mesh structures and the reconstruction of continuous surfaces or curves from discrete sets of points. By computing the Voronoi diagram of a point set, a set of polygonal or triangular meshes can be generated that approximate the surface shape of the original point set.

[0028] OccMap is a type of occupancy raster map that divides a 2D plane into small grids of equal length and width based on resolution, and fills in grid information according to whether a grid is occupied. In this scheme, if a lane line or road boundary exists at a certain location, it is considered an occupied grid.

[0029] ESDF (Euclidean Signed Distance Field) is an OccMap that stores the signed Euclidean distance from each cell to the nearest obstacle. The "signed" aspect refers to whether the object is outside (+) or inside (-), which is unnecessary in this design and therefore not implemented. In other words, this program actually implements EDF; ESDF is simply the more common terminology.

[0030] COIF (Closest Obstacle Index Field) is a byproduct of ESDF, recording the coordinates of the nearest obstacle.

[0031] In existing technologies, the rapid development of autonomous driving technology has placed core demands on road environment perception and map building, requiring high precision, real-time performance, and dynamic adaptability. Currently, environmental perception in autonomous driving systems mainly relies on two technical approaches: one is to achieve localization and path planning by matching real-time sensor data with pre-collected high-precision maps; the other is to adopt a map-less or map-sparse mode, relying on real-time sensor perception to build an environmental model.

[0032] While high-precision map-based technologies can provide centimeter-level road geometry and semantic information, they suffer from high data collection and update costs and long cycles. They also struggle to handle dynamic scenarios such as temporary construction, lane adjustments, and unexpected obstacles, leading to a disconnect between map timeliness and the actual road environment. Furthermore, high-precision maps have limited coverage, rendering them ineffective in uncollected areas or remote sections, thus restricting the adaptability of autonomous driving.

[0033] In map-less or map-sparse modes, existing technologies mostly rely on sensors such as LiDAR and cameras to perceive obstacles and passable areas in real time, and construct environmental models using methods such as grid maps, occupancy grids, or topology maps. However, these methods generally have the following drawbacks: First, they can only generate macroscopic boundaries of passable areas and cannot accurately delineate lane-level structures (such as lane centerlines, boundary lines, and widths), making it difficult for vehicles to make refined lane keeping and lane changing decisions; second, the construction of topology relationships often depends on random sampling or simple geometric partitioning, which can easily lead to path redundancy and missing key nodes, affecting the efficiency and safety of path planning; third, they lack the ability to model lane attributes (such as left and right neighbor relationships, steering constraints, and dynamic width adaptation), making it difficult to meet the navigation needs of complex intersections and multi-lane merging scenarios.

[0034] Furthermore, existing technologies often require re-executing the entire environmental modeling process when dealing with dynamic environmental changes, resulting in insufficient real-time performance and delays in vehicle response to sudden road conditions. Therefore, how to construct lane-level maps in real time, including accurate lane attributes, dynamic topological relationships, and environmental adaptability, without relying on high-precision maps, has become a key technical challenge for improving the robustness and scenario versatility of autonomous driving systems.

[0035] This application provides a solution for constructing lane maps based on the Veno map algorithm and transforming them into accurate lane-level maps. This application only requires the driving environment information to include all boundary lines necessary for lane formation, and uses only navigation as the direction of travel, without relying on the lane's shape, thus providing a more general and powerful solution. All attributes of the lanes ultimately generated in this application will be complete, including left / right boundaries and left / right neighbors, which provides greater convenience for downstream modules such as prediction and planning.

[0036] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or map building device capable of performing the above functions. The following description uses a map building device as an example to illustrate this embodiment and the subsequent embodiments.

[0037] Based on this, the embodiments of this application provide a map construction method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the map construction method of this application.

[0038] In this embodiment, the map construction method includes steps S10 to S30: Step S10: Obtain the navigation route information of the target vehicle, wherein the navigation route information is the preset driving path information of the target vehicle; For example, navigation route information refers to the driving route information of vehicle path planning provided through cloud navigation services, etc. The navigation route information includes the driving direction planned by the target vehicle and can be used for lane matching.

[0039] Step S20: Based on the navigation route information, the exit lane and entry lane of the target vehicle are matched in a pre-constructed lane map. The lane map is constructed based on driving environment information within a preset range of the target vehicle's location. The driving environment information includes the boundary lines of each lane. The exit lane includes the lanes that the target vehicle can currently exit from. The entry lane includes the lanes that the target vehicle can enter. For example, driving environment information refers to road environment data within a preset range of the target vehicle's location output by the vehicle's perception module. The preset range can be set according to actual conditions and is not specifically limited here. The driving environment information includes the boundary lines of each lane. Optionally, in this embodiment, the driving environment information requires an outer polygon containing the boundary lines for constructing the lane map. The perception module mainly includes sensors such as LiDAR, cameras, and millimeter-wave radar, used to perceive and acquire driving environment information in real time.

[0040] Reference Figure 2 , Figure 2 This is a schematic diagram of the overall process according to an embodiment of this application, such as... Figure 2 As shown, in this embodiment of the application, the boundary line can be selected and cut according to the driving environment information, and then a Veno diagram can be generated according to the boundary line. The intersection point in the Veno diagram is determined, and the connectivity between lanes is represented by a point-edge model. Here, the point is the intersection point of the lane line (such as the intersection point), and the edge is the lane connection line, which includes lane direction, steering constraints and adjacency relationship (such as left and right lane numbers).

[0041] For example, effective connections are filtered by calculating the distance between lane centerline endpoints and the cosine similarity of lane feature vectors, and finally fused into a topological matrix. Based on the lane geometric and semantic features in the driving environment information, the logical structure of the lane network is constructed through a topological inference algorithm, providing a road skeleton for downstream lane matching.

[0042] In this embodiment, the position of the target vehicle in the lane map is located, and the road segment corresponding to the position of the target vehicle in the lane map is taken as the exit lane. Furthermore, based on the driving direction as the core basis, and combined with the road segment where the target vehicle is currently located in the pre-constructed lane map, an entry lane that conforms to the driving direction and can be safely entered by the target vehicle is determined.

[0043] For example, the step of matching the exit lane and entry lane of the target vehicle in a pre-built lane map based on the navigation route information includes: The navigation route information is matched with the lane map by at least one of geometric matching, lane attribute matching, traffic condition matching, and dynamic environment matching to determine the entry lane and exit lane.

[0044] It should be noted that the geometric matching includes matching the user's location in the lane map with the lane location in the navigation route information; the lane attribute matching includes matching the lane type in the lane map with the lane type in the navigation route information; the traffic condition matching filters the lanes in the lane map based on the traffic conditions in the navigation route information; and the dynamic environment matching filters the lanes in the lane map based on real-time traffic information in the navigation route information. For example, user navigation will issue a navigation route as the main direction. In this embodiment, the lane map is matched with the navigation route to select a suitable road. The matching rules are mainly based on the geometric relationship between the user's location and the navigation route, combined with lane attribute information and road traffic conditions. Specifically, based on the user's location, a projection is made onto the navigation route, the closest point between the user's current location and the navigation route is calculated, and the user is classified into an entry lane and an exit lane based on the projection result.

[0045] For example, geometric matching primarily calculates the user's projection point on the navigation route based on the user's current location coordinates. A distance threshold is used to determine whether the user is approaching a lane of the navigation route; if the projection point is too far from the navigation route, the user is considered to have deviated from the navigation route.

[0046] For example, lane attribute matching primarily matches the lane type, such as main lane, auxiliary road, ramp, etc., to ensure that the navigation route is consistent with the lane type. Lane directionality can also be considered to ensure that the user's driving direction is consistent with the navigation route. For example, traffic condition matching includes checking the traffic conditions of lanes, such as whether there are closures, construction, or traffic restrictions, and prioritizing lanes with good traffic conditions for matching.

[0047] For example, dynamic environment matching primarily considers real-time traffic information, such as congestion and accident information, and dynamically adjusts the matching results. If a lane on the navigation route is severely congested, an adjacent alternative lane can be selected.

[0048] This embodiment can achieve efficient matching between navigation routes and lane maps through the above matching rules, ensuring that users can drive smoothly according to the navigation route, while dynamically adapting to complex road environments.

[0049] Step S30: Perform road connection processing on the exit lane and the entry lane to obtain a target lane-level map.

[0050] For example, based on the lane map, each Vino edge can generate one or more lanes according to the lane width. When the lane lines are clear, the lane width is determined based on the left and right boundaries of the lane lines. When there are blurry lane lines or empty areas, a fixed lane width is used. In actual road conditions, there will be some narrow or wide roads, in which case the lane width is calculated based on the lane line information.

[0051] For example, in this embodiment of the application, the intersection points of the lane map are used to construct lane connections within the road. In order to construct an accurate lane-level map, the lane centerline needs to undergo post-processing such as smoothing, extension, supplementing lane boundaries, and left and right neighbor attributes, and then the correct connections are given for different roads, thereby generating an accurate target lane-level map.

[0052] This embodiment, through the above-described scheme, specifically obtains the navigation route information of the target vehicle, wherein the navigation route information is the target vehicle's preset driving path information; based on the navigation route information, the exit lane and entry lane of the target vehicle are matched in a pre-constructed lane map, wherein the lane map is constructed based on driving environment information within a preset range of the target vehicle's location, the driving environment information including the boundary lines of each lane; the exit lane includes the lanes that the target vehicle can currently exit from; the entry lane includes the lanes that the target vehicle can enter; road connection processing is performed on the exit lane and the entry lane to obtain a target lane-level map. The lane map can be generated using only driving environment information. The exit lane and entry lane are determined in the lane map based on the navigation route information, and then road connection processing is performed, thereby constructing an accurate lane-level map. This improves the accuracy and stability of lane-level map construction without relying on high-precision maps, thereby enhancing the safety and scenario versatility of the autonomous driving system.

[0053] Based on the first embodiment of this application, a second embodiment of this application is proposed. In this second embodiment, content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. Based on this, please refer to... Figure 3 The lane map is constructed according to the following steps: Step S201: Obtain driving environment information within a preset range of the target vehicle's location; Step S202: Generate rasterized data based on the lane boundary lines in the driving environment information; For example, in this embodiment, the driving environment information within a preset range of the target vehicle's location is first obtained. This driving environment information includes lane boundary lines, which are markers used to define the physical extent of lanes and separate different lanes. Further, through sensor calibration and spatiotemporal synchronization, multi-source data such as LiDAR point clouds and camera images are fused to generate a three-dimensional point cloud of the driving environment. Then, the point cloud is projected onto a two-dimensional plane using voxel gridding to mark obstacle grids. Finally, the Euclidean distance (ESDF) from each passable grid to the nearest obstacle and the obstacle coordinates (COIF) are calculated to quantify grid safety and spatial correlation. This solves the sparsity and noise problems of the original sensing data, achieves a dense representation of the environment through gridding, and provides distance and coordinate quantification basis for Vinograph extraction through ESDF / COIF.

[0054] Step S203: Extract the rasterized data to generate a Venn diagram; Reference Figure 4 , Figure 4 This is a schematic diagram of a Vinio diagram according to an embodiment of this application, such as... Figure 4 As shown, the steps for establishing a Vinio diagram in this embodiment include: (1) Use OccMap to construct ESDF and COIF; (2) Extract the Voronoi Map. For each point, determine whether there is another 4-connected point such that its nearest obstacle is different from the nearest obstacle of this point. If true, set the value to 1. (3) To refine the Voronoi Map, the Zhang-Suen refinement algorithm can be used.

[0055] For example, obstacle center points are extracted from rasterized data as seed points for the Veno map; then, for each passable raster, the coordinates (COIF) of its four-connected neighborhood (up, down, left, right) are checked to see if they are the same as its own. If they are different, the raster lies on the perpendicular bisector of the two obstacles and is marked as a "Veno edge" (value = 1); finally, a binary Veno map is formed. The physical shape information (obstacles) of the road and lane boundaries is used as input to the Veno map, similar to... Figure 4 The perpendicular bisectors generated from the discrete points in the data can be understood as the initial lane centerlines.

[0056] Step S204: Refine the Vinio map and merge the branch points to obtain the lane map.

[0057] The steps of refining the Vinio map and merging branch points to obtain the lane map include: The Venn diagram is refined to generate lane centerlines; and the branch points in the Venn diagram are merged to generate at least one intersection point. The lane map is constructed based on the lane centerline and the at least one intersection point.

[0058] For example, in the embodiments of this application, the Vinno map can be thinned using a thinning algorithm, including the Zhang-Suen algorithm. The Zhang-Suen algorithm iteratively thins the multi-pixel-wide Vinno edges into continuous skeleton lines of single-pixel width, eliminating redundant pixels and preserving topological connectivity.

[0059] For example, the at least one intersection point includes a first type of intersection point and / or a second type of intersection point, and the step of merging the branch points in the Venn diagram to generate at least one intersection point includes: The degree of each branch point is determined based on the number of connected edges at each branch point in the Vinograph. Designate the branch points with a degree of 1 as leaf nodes and the branch points with a degree greater than 1 as non-leaf nodes. Adjacent points whose distance to the leaf node is less than a first preset distance threshold are merged to obtain the first type of intersection point; and adjacent points whose distance to the non-leaf node is less than a second preset distance threshold are merged to obtain the second type of intersection point.

[0060] For example, based on the single-pixel skeleton lines of the refined Veno diagram, the number of valid connecting edges is counted through an 8-neighborhood search (for each branch point, check whether its 8 adjacent pixels in each direction are skeleton lines). If an adjacent pixel is a skeleton line and the angle between its connecting direction and the current point is >30° (to avoid counting the same edge repeatedly), it is counted as a valid edge, and the sum of the edge counts is the degree of the node.

[0061] Reference Figure 5 , Figure 5 This is a schematic diagram of the intersection point according to an embodiment of this application, such as... Figure 5 As shown, nodes are classified based on their calculated degree values. Nodes with a degree of 1 are defined as leaf nodes (connected by only one edge, with no branching ability); nodes with a degree of ≥2 are defined as non-leaf nodes (possessing multi-directional connectivity), and non-leaf nodes can also be called branch points. Isolated nodes (noise) with a degree of 0 are directly discarded and do not participate in subsequent processing.

[0062] For example, the "K-nearest neighbor search + cluster merging" algorithm is adopted. First, spatial indexing is performed on all leaf nodes using a KD-tree (K-Dimensional Tree) to quickly find the K-nearest neighbors (e.g., K=5) of each leaf node; then, the Euclidean distance between the leaf node and its nearest neighbor is calculated. If the distance is ≤ a first preset distance threshold (e.g., 4.6 meters), they are grouped into the same cluster; finally, the geometric center of each cluster is calculated as the first type of intersection point, and the original leaf node is deleted.

[0063] For example, the topological distance (number of edges) between non-leaf nodes is calculated by breadth-first search (BFS), and the nearest neighbor nodes with a topological distance of less than or equal to 2 are filtered out; then their geometric distance is calculated, and if it is less than or equal to a second preset distance threshold (such as 100 meters), they are merged into the same cluster; finally, the second type of intersection point is determined by voting (such as taking the turning center point with the highest frequency).

[0064] For example, in this embodiment of the application, branch points (nodes with degree ≥ 2) are identified through 8-neighborhood analysis, leaf nodes (degree = 1) and non-leaf nodes (degree ≥ 2) are distinguished, and the geometric center is calculated as the intersection point after merging according to the threshold. This can preserve key connectivity relationships and eliminate redundant nodes, providing efficient and reliable map support for real-time path planning of autonomous driving.

[0065] This embodiment, through the above-described scheme, specifically obtains driving environment information within a preset range of the target vehicle's location; generates rasterized data based on the lane boundary lines in the driving environment information; extracts data from the rasterized data to generate a Venn diagram; and refines and merges branch points on the Venn diagram to obtain the lane map. By generating the Venn diagram using only the boundary lines required to form lanes contained in the driving environment information, and refining and merging branch points on the Venn diagram, key connectivity relationships can be preserved, and redundant nodes can be eliminated, providing an efficient and reliable lane map for real-time path planning in autonomous driving.

[0066] Based on any of the foregoing embodiments of this application, a third embodiment of this application is proposed. In this third embodiment, content that is the same as or similar to that of Embodiment 1 above can be referred to the above description and will not be repeated hereafter. Based on this, please refer to... Figure 6 Step S30 also includes steps S301 to S303: Step S301: Determine the center lines of the exit lane and the entry lane respectively according to the lane map; Reference Figure 7 , Figure 7 This is a schematic diagram illustrating driving environment information according to an embodiment of this application, such as... Figure 7As shown, driving environment information can provide all the boundary lines of the lane, forming an outer polygon. Rasterized data is generated based on the driving environment information. The rasterized data is extracted to generate a Venn diagram. The Venn diagram is then refined and its branch points are merged to obtain a lane map, which includes the generated lane centerline.

[0067] Reference Figure 8 , Figure 8 This is a schematic diagram of a lane map according to an embodiment of this application, such as... Figure 8 As shown, the pink lines in the figure represent the Venn diagram of the topology-level lanes. In this embodiment, the sequence of intersection points corresponding to the target lane (e.g., P0→P1→P2) is first extracted from the lane map; then, the shortest path is searched using Dijkstra's algorithm to ensure that the intersection points are connected in an orderly manner according to the navigation direction; finally, the intersection points are connected sequentially by straight line segments to form a polygonal lane centerline. This transforms the topology-level abstract path into quantifiable geometric lines, providing a continuous curve basis for subsequent smoothing and boundary generation.

[0068] Step S302: Smooth and extend the center lines of the lanes corresponding to the exit lane and the entry lane respectively, and supplement the lane attributes to generate the processed entry lane and exit lane. It should be noted that the lane attributes of the entering lane include the boundary line forming the entering lane and / or the lanes adjacent to the entering lane, and the lane attributes of the exit lane include the boundary line forming the exit lane and / or the lanes adjacent to the exit lane. For example, based on the lane map, each Vino edge can generate one or more lanes according to the lane width. When the lane lines are clear, the lane width is determined based on the left and right boundaries of the lane lines. When there are blurry lane lines or empty areas, a fixed road width (e.g., 3.5m) is used. In actual road conditions, there may be narrow or wide roads, in which case the lane width is calculated based on the lane line information. The intersection points of the lane map are used to construct the lane connections within the road. To construct an accurate lane-level map, the lane centerline needs to undergo post-processing such as smoothing, extension, supplementing lane boundaries, and adjusting left and right neighbor attributes to obtain the processed entering and exit lanes.

[0069] Step S303: Perform road connection processing on the processed inbound and outbound lanes to obtain the target lane-level map.

[0070] For example, the step of performing road connection processing on the processed inbound and outbound lanes to obtain the target lane-level map includes: Extract the lane lines corresponding to the processed inbound and outbound lanes; The lane lines to be connected are selected from the lane lines corresponding to the processed exit lanes. The lane lines to be connected and the lane lines corresponding to the processed entering lanes are topologically connected to obtain the target lane-level map.

[0071] Reference Figure 9 , Figure 9 This is a schematic diagram of a left-turn scenario according to an embodiment of this application, such as... Figure 9 As shown, the connection rules between roads are illustrated using left turns as an example in this embodiment. Figure 9 The two rectangles in the diagram represent the constructed roads. The dashed lines within these rectangles represent the center lines of the constructed roads, i.e., the lanes. For the two left-turn roads crossing intersections, and their respective lanes, connecting lines are needed to ensure the continuity of the road structure. For ease of description, the two roads in the diagram are defined as the exit lanes, i.e., parent roads. Figure 9 The first road where the target vehicle is located, and the other is the child road, which is the lane the vehicle is entering. Figure 9 The road in front of the target vehicle is defined as follows: lanes within the parent road are defined as parentlanes, and lanes within the child road are defined as childlanes. The specific process includes: 1. Retrieve parentlane and childlane information from parentroad and childroad; 2. When connecting, it is necessary to determine whether the connecting and outgoing lanes are non-motorized vehicle lanes (e.g., lanes for motorized vehicles). Figure 9 The rightmost yellow lane in the childroad lane is the non-motorized vehicle lane, so it needs to be shielded when connecting and cannot be connected outwards. 3. Based on point 2, the connection process needs to be handled in two cases: Case a. All lanes in the parent road are non-motorized vehicle lanes; Case b. The parent road has some non-motorized vehicle lanes or none at all. 4. For case a in 3: In order to ensure that vehicles have a way to go, the non-motorized vehicle lane of the first parentlane from left to right in the parentroad will be covered and used as the last parentlane to be connected; For case b: Among the lanes that are not non-motorized vehicle lanes, the rightmost lane with left turn attribute will be found and used as the last parentlane to be connected. 5. For the found parentlane, find the corresponding childlane in the childroad based on its position (index) in the road, and perform topological connection; 6. Finally, for the remaining unconnected childlanes, traverse them and connect them with the found parentlanes.

[0072] Reference Figure 10 , Figure 10 This is a schematic diagram of a target lane-level map according to an embodiment of this application, such as... Figure 10 As shown, the orange and yellow lines are precise lane centerlines generated by the algorithm based on the Vinyasa diagram, thus constructing an accurate target lane-level map. The discrete Vinyasa diagram algorithm in this embodiment can be completed in θ(N), where N is the number of grid cells, which is Σ(road area) / (resolution^2). For example, in a typical scenario of a 120m*20m road with a resolution of 0.5m, the number of grid cells is 9600. Even without sufficient compilation options, this step can be completed within 20ms on a modern computer. Even on embedded platforms such as TDA4, it can be accelerated using parallelism to achieve real-time performance. Therefore, the map construction scheme in this embodiment has low time consumption and improves the efficiency of lane-level map construction.

[0073] This embodiment, through the above-described scheme, specifically determines the lane centerlines corresponding to the exit lane and the entry lane based on the lane map; smooths and extends the lane centerlines corresponding to the exit lane and the entry lane, and supplements lane attributes to generate processed entry lanes and exit lanes; performs road connection processing on the processed entry lanes and exit lanes to obtain the target lane-level map. This achieves improved accuracy, stability, and efficiency of lane-level map construction without relying on high-precision maps, thereby enhancing the safety and scenario applicability of the autonomous driving system.

[0074] Based on any of the foregoing embodiments of this application, a fourth embodiment of this application is proposed. In this fourth embodiment, content that is the same as or similar to that of Embodiment 1 above can be referred to the above description and will not be repeated hereafter. Based on this, the step of selecting the lane line to be connected from the lane lines corresponding to the processed exit lane includes: When there is one driving lane in the processed exit lane, the lane line corresponding to the processed exit lane is used as the lane line to be connected. When there are at least two driving lanes in the processed exit lane, determine the lane type of each driving lane; If there is a driving lane of the type of motor vehicle lane, then determine the driving direction indicator associated with the driving lane of the type of motor vehicle lane, wherein the driving direction indicator is used to indicate the direction in which the target vehicle can travel in the driving lane; Based on the driving direction indicator, the lane line to be connected is selected from the lane lines corresponding to the driving lanes belonging to the motor vehicle lane type.

[0075] It should be noted that lane types include motor vehicle lanes and non-motor vehicle lanes. The driving direction markings are used to indicate the direction that the target vehicle can travel in the driving lane, such as markings for going straight, turning left, and turning right.

[0076] In one embodiment, when there is one driving lane in the processed exit lane, the lane line corresponding to the processed exit lane is directly used as the lane line to be connected.

[0077] In another embodiment, when the processed exit lane has multiple driving lanes, lanes that do not conform to the vehicle attributes need to be excluded first, and then the lane lines to be connected are determined according to the navigation route direction. The vehicle attributes include motor vehicles and non-motor vehicles. This embodiment uses motor vehicles as an example for illustration. Specifically, when the processed exit lane has at least two driving lanes, the lane type of each driving lane is determined to exclude non-motor vehicle driving lanes and retain motor vehicle driving lanes. Further, if there is a driving lane of the motor vehicle type, the driving direction identifier associated with the driving lane of the motor vehicle type is determined, and then, based on the driving direction identifier and the navigation direction of the target vehicle in the navigation route information, the lane lines to be connected are selected from the lane lines corresponding to the driving lanes of the motor vehicle type. For example, refer to Figure 9 If the target vehicle's navigation direction is left turn, select the lane line corresponding to the driving lane marked as left turn as the lane line to be connected.

[0078] This embodiment, through the above-described scheme, selects the lane lines to be connected from the lane lines corresponding to the driving lane by combining the lane type and driving direction indicators, ensuring that the driving path is consistent with the navigation plan. Subsequently, the lane lines to be connected and the processed lane lines corresponding to the entering lane are topologically linked to obtain a target lane-level map. This improves the accuracy, stability, and efficiency of lane-level map construction without relying on high-precision maps, thereby enhancing the safety and scenario applicability of the autonomous driving system.

[0079] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the map construction method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0080] This application also provides a map building apparatus, please refer to... Figure 11 The map building apparatus includes: The acquisition module 10 is used to acquire navigation route information of the target vehicle, wherein the navigation route information is the preset driving path information of the target vehicle; The matching module 20 is used to match the exit lane and entry lane of the target vehicle in a pre-constructed lane map based on the navigation route information. The lane map is constructed based on driving environment information within a preset range of the target vehicle's location. The driving environment information includes the boundary lines of each lane. The exit lane includes the lanes that the target vehicle can currently exit from. The entry lane includes the lanes that the target vehicle can enter. The connection module 30 is used to perform road connection processing on the exit lane and the entry lane to obtain a target lane-level map.

[0081] In one embodiment, the map building apparatus includes: The driving environment information acquisition module is used to acquire driving environment information within a preset range of the target vehicle's location; The first generation module is used to generate rasterized data based on the lane boundary lines in the driving environment information. The second generation module is used to extract the rasterized data and generate a Venn diagram; The merging module is used to refine the Vinio map and merge branch points to obtain the lane map.

[0082] In one embodiment, the merging module includes: A refinement unit is used to refine the Vinio diagram and generate lane center lines; A merging unit is used to merge the branch points in the Venn diagram to generate at least one intersection point; A constitutive unit is used to construct the lane map based on the lane centerline and the at least one intersection point.

[0083] In one embodiment, the merging unit includes: Select a component to determine the degree of each branch point based on the number of connected edges at each branch point in the Vinograph; Determine the component to designate the corresponding branch point with a degree of 1 as a leaf node and the corresponding branch point with a degree greater than 1 as a non-leaf node; The merging component is used to merge adjacent points whose distance to the leaf node is less than a first preset distance threshold to obtain the first type of intersection point; and to merge adjacent points whose distance to the non-leaf node is less than a second preset distance threshold to obtain the second type of intersection point.

[0084] In one embodiment, the matching module includes: The first determining unit is configured to perform at least one of the following: geometric matching, lane attribute matching, traffic condition matching, and dynamic environment matching, on the navigation route information and the lane map, to obtain the exit lane and the entry lane; wherein, the geometric matching includes matching the user's location in the lane map with the lane location in the navigation route information; the lane attribute matching includes matching the lane type in the lane map with the lane type in the navigation route information; the traffic condition matching filters the lanes in the lane map according to the traffic conditions in the navigation route information; and the dynamic environment matching filters the lanes in the lane map according to the real-time traffic information in the navigation route information.

[0085] In one embodiment, the connection module includes: The second determining unit is used to determine the lane center lines corresponding to the exit lane and the entry lane respectively, based on the lane map. The processing unit is used to smooth and extend the center lines of the lanes corresponding to the exit lane and the entry lane respectively, and supplement the lane attributes to generate the processed entry lane and exit lane. The lane attributes of the entry lane include the boundary line forming the entry lane and / or the lane adjacent to the entry lane, and the lane attributes of the exit lane include the boundary line forming the exit lane and / or the lane adjacent to the exit lane. The connection unit is used to perform road connection processing on the processed inbound and outbound lanes to obtain the target lane-level map.

[0086] In one embodiment, the connection unit includes: The lane line extraction unit is used to extract the lane lines corresponding to the processed inbound and outbound lanes. The selection unit is used to select the lane line to be connected from the lane lines corresponding to the processed exit lane. The topology connection unit is used to connect the lane lines to be connected and the lane lines corresponding to the processed entering lanes in a topology to obtain the target lane-level map.

[0087] In one embodiment, the selection unit includes: The processing subunit is used to take the lane line corresponding to the processed exit lane as the lane line to be connected when there is one driving lane in the processed exit lane. The first determining subunit is used to determine the lane type of each of the driving lanes when there are at least two driving lanes in the processed exit lane. The second determining subunit is used to determine the driving direction identifier associated with the driving lane belonging to the motor vehicle lane type if there is a driving lane of the motor vehicle lane type, wherein the driving direction identifier is used to indicate the direction in which the target vehicle can travel on the driving lane; A sub-unit is selected to select the lane line to be connected from the lane lines corresponding to the driving lanes belonging to the motor vehicle lane type, based on the driving direction indicator.

[0088] The map building apparatus provided in this application, employing the map building method in the above embodiments, can solve the technical problems of map building. Compared with the prior art, the beneficial effects of the map building apparatus provided in this application are the same as those of the map building method provided in the above embodiments, and other technical features in the map building apparatus are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0089] This application provides a map building device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the map building method in Embodiment 1 above.

[0090] The following is for reference. Figure 12 The diagram illustrates a structural schematic of a map building device suitable for implementing embodiments of this application. The map building device in these embodiments may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 12 The map building device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0091] like Figure 12As shown, the map building device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the map building device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the map-building device to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows map-building devices with various systems, it should be understood that implementing or having all of the systems shown is not required. More or fewer systems may be implemented alternatively.

[0092] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0093] The map building device provided in this application, employing the map building method in the above embodiments, can solve the technical problems of map building. Compared with the prior art, the beneficial effects of the map building device provided in this application are the same as those of the map building method provided in the above embodiments, and other technical features in this map building device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0094] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0095] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0096] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the map construction method in the above embodiments.

[0097] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0098] The aforementioned computer-readable storage medium may be included in the map building device or may exist independently without being assembled into the map building device.

[0099] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by a map-building device, the map-building device: acquires navigation route information of a target vehicle, wherein the navigation route information is a preset driving path information of the target vehicle; matches the exit lane and entry lane of the target vehicle in a pre-constructed lane map based on the navigation route information, wherein the lane map is constructed based on driving environment information within a preset range of the target vehicle's location, the driving environment information including the boundary lines of each lane; the exit lane includes the lanes that the target vehicle can currently exit from; the entry lane includes the lanes that the target vehicle can enter; performs road connection processing on the exit lane and the entry lane to obtain a target lane-level map. This method generates a lane map using only driving environment information, determines the exit lane and entry lane in the lane map based on the navigation route information, and then performs road connection processing, thereby constructing an accurate lane-level map. This improves the accuracy and stability of lane-level map construction without relying on high-precision maps, thereby enhancing the safety and scenario versatility of the autonomous driving system.

[0100] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0101] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0102] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0103] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described map construction method, thereby solving the technical problem of map construction. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the map construction method provided in the above embodiments, and will not be repeated here.

[0104] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the map construction method described above.

[0105] The computer program product provided in this application can solve the technical problem of map construction. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the map construction method provided in the above embodiments, and will not be repeated here.

[0106] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A map construction method characterized by comprising: The method comprises: obtaining navigation route information of a target vehicle, wherein the navigation route information is preset driving path information of the target vehicle; matching the target vehicle's exiting lane and entering lane in a pre-constructed lane map according to the navigation route information, wherein the lane map is constructed according to driving environment information within a preset range of the target vehicle's location, the driving environment information comprising various lane boundary lines; the exiting lane comprises a lane that the target vehicle can currently exit; and the entering lane comprises a lane that the target vehicle can enter; performing road connection processing on the exiting lane and the entering lane to obtain a target lane-level map.

2. The map construction method according to claim 1, wherein The lane map is constructed according to the following steps: obtaining driving environment information within a preset range of the target vehicle's location; generating rasterized data according to various lane boundary lines in the driving environment information; extracting the rasterized data to generate a V-Map; performing refinement processing and branch point merging on the V-Map to obtain the lane map.

3. The map construction method according to claim 2, wherein The step of performing refinement processing and branch point merging on the V-Map to obtain the lane map comprises: performing refinement processing on the V-Map to generate a lane center line; and merging various branch points in the V-Map to generate at least one intersection point; constructing the lane map according to the lane center line and the at least one intersection point.

4. The map construction method according to claim 3, wherein The at least one intersection point comprises a first type of intersection point and / or a second type of intersection point, and the step of merging various branch points in the V-Map to generate at least one intersection point comprises: determining the degrees corresponding to various branch points according to the number of connection edges of the branch points in the V-Map; regarding branch points with a corresponding degree of 1 as leaf nodes and regarding branch points with a corresponding degree greater than 1 as non-leaf nodes; merging adjacent points with a distance less than a first preset distance threshold from the leaf nodes to obtain the first type of intersection point; and merging adjacent points with a distance less than a second preset distance threshold from the non-leaf nodes to obtain the second type of intersection point.

5. The map construction method according to claim 1, wherein The step of matching the target vehicle's exiting lane and entering lane in a pre-constructed lane map according to the navigation route information comprises: performing at least one of geometric matching, lane attribute matching, traffic condition matching, and dynamic environment matching on the navigation route information and the lane map to obtain the exiting lane and the entering lane; wherein the geometric matching comprises matching a user location in the lane map with a lane location in the navigation route information, the lane attribute matching comprises matching a lane type in the lane map with a lane type in the navigation route information, the traffic condition matching comprises screening a lane in the lane map according to a traffic condition in the navigation route information, and the dynamic environment matching comprises screening a lane in the lane map according to real-time traffic information in the navigation route information.

6. The map construction method according to claim 1, wherein The step of performing road connection processing on the exiting lane and the entering lane to obtain a target lane-level map comprises: According to the lane map, the lane center lines corresponding to the exiting lane and the entering lane are determined respectively; The lane center lines corresponding to the exiting lane and the entering lane are respectively subjected to smoothing and extension processing, and lane attributes are supplemented, to generate processed exiting lane and entering lane, wherein the lane attributes of the entering lane include boundary lines forming the entering lane and / or lanes adjacent to the entering lane, and the lane attributes of the exiting lane include boundary lines forming the exiting lane and / or lanes adjacent to the exiting lane; The processed exiting lane and the processed entering lane are subjected to road connection processing to obtain the target lane-level map.

7. The map construction method according to claim 6, wherein The step of obtaining the target lane-level map by performing road connection processing on the processed exiting lane and the processed entering lane comprises: extracting lane lines corresponding to the processed exiting lane and the processed entering lane; selecting lane lines to be connected from the lane lines corresponding to the processed exiting lane; topologically connecting the lane lines to be connected and the lane lines corresponding to the processed entering lane to obtain the target lane-level map.

8. The map construction method according to claim 7, wherein The step of selecting lane lines to be connected from the lane lines corresponding to the processed exiting lane comprises: when the processed exiting lane has one driving lane, the lane lines corresponding to the processed exiting lane are selected as the lane lines to be connected; when the processed exiting lane has at least two driving lanes, the lane types of the driving lanes are determined; if there is a driving lane whose lane type belongs to a motor lane type, a driving direction identifier associated with the driving lane belonging to the motor lane type is determined, wherein the driving direction identifier is used to indicate a drivable direction of the target vehicle on the driving lane; the lane lines to be connected are selected from the lane lines corresponding to the driving lane belonging to the motor lane type according to the driving direction identifier.

9. A map construction apparatus characterized by comprising: The device comprises: an acquisition module configured to acquire navigation route information of a target vehicle, wherein the navigation route information is preset driving path information of the target vehicle; a matching module configured to match an exiting lane and an entering lane of the target vehicle in a lane map constructed in advance according to the navigation route information, wherein the lane map is constructed according to driving environment information within a preset range of a location of the target vehicle, the driving environment information includes various lane boundary lines, the exiting lane includes a lane currently drivable by the target vehicle, and the entering lane includes a lane drivable by the target vehicle; a connection module configured to perform road connection processing on the exiting lane and the entering lane to obtain a target lane-level map.

10. A map construction device characterized by comprising: The device comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the map construction method according to any one of claims 1 to 8.

11. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program. The computer program is executed by the processor to implement the steps of the map construction method in any one of claims 1 to 8.

Images